Run This Ai
EN DE

TabbyAPI

Lightweight OpenAI-compatible ExLlamaV2 API server

★ 1,257 GitHub AGPL-3.0 llminferenceexllamaapi-serveropenai-compatibleself-hostedquantization LLM & Chat

Überblick

TabbyAPI is the official API backend server for ExLlamaV2 and ExLlamaV3, providing a lightweight, fast FastAPI-based application for generating text using large language models. It offers a fully OpenAI-compatible API with support for loading/unloading models, HuggingFace model downloading, tool calling, LoRA adapters, and more. TabbyAPI publishes a CUDA-ready Docker image via GitHub Container Registry, making deployment straightforward on GPU-equipped servers. It is designed for hobbyist and small-scale use, with easy setup via Docker or manual Python installation.

Anforderungen

Min vCPU
2
Min RAM
4096 MB
Min Disk
10 GB
Rec vCPU
8
Rec RAM
16384 MB
Rec Disk
20 GB

Empfohlener VPS

Hostinger · KVM 8

8 vCPU · 32256 MB · 400 GB

$20.00
Zum Anbieter

Affiliate-Hinweis

Docker Compose

# Generated by Run This Ai — docker-compose.yml
services:
  tabbyapi:
    image: ghcr.io/theroyallab/tabbyapi:latest
    restart: unless-stopped
    ports:
      - 8080:8080
    volumes:
      - ./data/tabbyapi:/data

Bester VPS für TabbyAPI →

Verwandte Tools

Anleitungen & Artikel